Everybody knows that city of Houston elections are fairly low-turnout affairs. The general perception – and it’s one that I’ve echoed as well – is that the elections are dominated by the same voters, year after year.
What I haven’t seen are the numbers to back up those assertions. With all that’s on the line in this year’s election, I thought this might be a good time to try and figure some of this stuff out. So I got the election files from the Harris County Clerk for all of the odd-numbered years going back to 2003, and played around with them to see what I could learn.
Now, we know that the bulk of Houston voters are in Harris County, but not all of them. There are a couple of thousand voters in Fort Bend County, and a couple of dozen voters in Montgomery County. I didn’t try to get the data for these voters in these elections. They represent maybe two percent of the total each year, and I just didn’t feel like the effort to include that data was worthwhile. So when you see me tossing around total turnout numbers in this post, bear in mind I’m talking about Harris County turnout for the city of Houston.
The first question I wanted to answer was “Who are the truly hardcore voters in Houston, the ones who come out every election without fail?” My sense going into this was that there might be fewer of these people than one might think. The answer is that 36,036 City of Houston voters have participated in every November election since 2003. That’s 20.7% of the 174,132 Harris County voters who cast a ballot in Houston in 2013.
I don’t know what kind of number I was expecting, but I wasn’t terribly surprised by this. Houston is a dynamic city. People move in and move out, sometimes to or from the suburbs in Harris County and elsewhere, sometimes to and from other cities or states, or countries. We have a significant population of children, and some of them turn 18 every year. We also have an electorate that skews old – more on that in the next post from this series – and every year some of them age out, which is the poli-sci way of saying “die”. A lot of this can and will happen over a ten-year span. The total number of voters in a given year may reliably be in a narrow band, but the names do indeed change over time.
What about the short term? The gold standard for voters to contact for candidate outreach is those who have voted in at least two of the last three elections. How many potential voters does that rule out, given what we now know about the amount of turnover in the electorate? Consider the 2013 election, which had a near identical voter total as the open 2009 election. 54,708 of 174,132 voters in 2013 had not voted in either of the 2011 or 2009 elections. That’s 31.4% of the total. The two-out-of-threes are a clear majority, but that’s still an awful lot of votes to leave on the table if you don’t try to find them.
This isn’t new. In 2009, 63,164 voters had not participated in 2003 – this is 36.1% of the 175,031 total voters. 58,973 2009 voters – 33.7% – had not voted in either 2005 or 2007. I feel pretty confident saying that when we look back on the 2015 election, we will find that something like 35% of the electorate was “new”. Given the past pattern of turnout being higher in years with high-profile referenda on the ballot, that’s likely to be an understatement.
Who are these “new” voters? As I’ve said before, some of them are new to Houston, and some of them are newly registered. Some of them have been here all along, and just hadn’t had a reason to come out to the ballot box before. How many of each there are, I couldn’t say. I can say that a candidate or campaign that isn’t trying to find and engage these voters is missing a significant opportunity. Especially in a year like this, that’s not a good idea.
Great data mining. I believe (though I am not positive) that I am one of those 36,000!
It’s possible that some voters have missed one or two elections since 2003 and still be considered a “committed” voter. For example, some individuals may have a health or family issue one year that prevents him or her from voting. Also, in some years a voter may make a conscious decision not to vote because he or she doesn’t believe his vote matters. I remember deciding not to vote during a single race runoff because the winner was already a foregone conclusion. I think a more interesting metric might be to compute the number who voted in 3 of the last 4 general elections — regardless of whether they failed to vote in 2013 or 2003.
By the way, I am assuming that this 36,000 number refers to individual voters regardless of how many times they have changed addresses. I think I have changed addresses 3 times since 2003.
Robert – There was a unique “voter ID number” field that I used as a join between election years. So yes, it should work regardless of how many times you changed address. I’ll check the data when I get the chance to ensure you’re in that “all six elections” table. Thanks!
Former Mayor Kathy Whitmire got me interested in voting in City of Houston elections, and I’ve voted in all of them since. If “age out” doesn’t get me, I’ll be voting in this one, too.
Proudly one of the 20.7% and will be taking my 5 year old daughter to the voting booth with me.
On a related note I wish the USA had $ fines for citizens who do not vote, like Australia
Pingback: Eye on Williamson » TPA Blog Round Up (August 24, 2015)
Pingback: Texas blog roundup for the week of August 24 – Off the Kuff